MONTE-CARLO SIMULATION OF MODELS FOR SINGLE POLYETHYLENE COILS

A variant of a simple sampling Monte Carlo technique is applied to study the configurational statistics of models for short isolated polyethylene chains, taking into account variations of the bond lengths, the bond angles, and the torsional angles as well as a Lennard-Jones (LJ) potential between nonbonded neighbors. The dependence of the mean-square end-to-end distance [R2], of the gyration radius [R(g)2], and of their ratio [R2]/[R(g)2] on temperature T and chain length N is studied. We also investigated the characteristic ratio C(infinity)(T) and distribution functions P(n)(l) and P(n)(theta), where l stands for the length of groups formed from n successive monomers and theta for the angle between two such successive groups. In a THETA-solvent or a dense melt the LJ interaction will be screened due to the polarizability of the solvent or due to other chains, respectively. We model this with a cutoff in the LJ interaction and show that all data depend very sensitively on the way in which the Lennard-Jones potential is truncated. Furthermore, we note that different potential models for polyethylene yield significantly different results. In view of these problems, it is not obvious that one can identify such configurations of isolated single chains with coil configurations in either THETA-solvent or a dense melt, respectively. Possible consequences for various approaches toward the chemically realistic modeling of polymeric materials are briefly discussed.